Skip to content

Latest commit

 

History

19 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

TinySteps Logo

TinySteps

AI-Powered Infant Growth & Development Monitoring System

A research-driven mobile healthcare platform that leverages Artificial Intelligence to support early childhood development through growth monitoring, Autism risk screening, maternal recovery assessment, and infant discomfort recognition.

Python FastAPI React Native Supabase TensorFlow License


TinySteps

TinySteps is an AI-powered mobile healthcare platform developed as our Final Year Research Project for the Bachelor of Science (Hons) in Information Technology specializing in Data Science at the Sri Lanka Institute of Information Technology (SLIIT).

The platform combines Artificial Intelligence, Machine Learning, Explainable AI, and Mobile Health technologies into a unified system that assists caregivers in monitoring infant growth and development while providing intelligent healthcare insights.


Features

  • Autism Spectrum Disorder (ASD) Risk Screening
  • Infant Growth Monitoring
  • Growth Prediction using LSTM
  • Maternal Recovery Assessment
  • Infant Cry & Discomfort Recognition
  • Explainable AI (SHAP & Grad-CAM)
  • Personalized Recommendations
  • WHO Growth Monitoring
  • Clinical Alert Generation
  • Daily Health Logging
  • Secure Authentication
  • REST API Architecture

System Overview

TinySteps consists of four integrated AI healthcare modules operating through a centralized FastAPI backend.

                          ┌─────────────────────────────┐
                          │   TinySteps Mobile App      │
                          │   React Native + Expo       │
                          └──────────────┬──────────────┘
                                         │
                                 HTTPS REST API
                                         │
                          ┌──────────────▼──────────────┐
                          │      FastAPI Backend        │
                          │ Authentication & Business   │
                          │ Logic + AI Service Layer    │
                          └──────────────┬──────────────┘
                                         │
               ┌─────────────────────────┼────────────────────────┐
               │                         │                        │
               ▼                         ▼                        ▼
       ┌──────────────┐         ┌────────────────┐      ┌────────────────┐
       │  Supabase DB │         │ Machine Learning│      │ Storage Layer │
       │ PostgreSQL   │         │ Model Services │      │ Images / Audio│
       └──────┬───────┘         └────────┬───────┘      └────────────────┘
              │                          │
              └──────────────┬───────────┘
                             │
      ┌──────────────────────┼─────────────────────────────┐
      ▼                      ▼                             ▼
┌───────────────┐   ┌────────────────┐          ┌──────────────────┐
│ ASD Detection │   │ Growth Module  │          │ Maternal Recovery│
│ VGGFace       │   │ LSTM           │          │ Random Forest    │
│ XGBoost       │   │ Random Forest  │          │ Ridge Regression │
│ ExplainableAI │   │ XGBoost        │          │ SHAP             │
└───────────────┘   └────────────────┘          └──────────────────┘
                             │
                             ▼
                  ┌────────────────────────┐
                  │ Infant Cry Recognition │
                  │ Gradient Boosting      │
                  │ MediaPipe              │
                  │ XGBoost Fusion         │
                  └────────────────────────┘

AI Components

Autism Spectrum Disorder Detection

  • Facial Image Analysis
  • Q-CHAT Assessment
  • Explainable AI
  • VGG-Face
  • XGBoost
  • Monte Carlo Late Fusion

Infant Growth Monitoring

  • Daily Growth Logging
  • Growth Prediction
  • LSTM Forecasting
  • WHO WAZ Monitoring
  • Random Forest
  • XGBoost
  • Clinical Rule Engine

Postpartum Maternal Recovery

  • Recovery Assessment
  • Personalized Recommendations
  • Explainable AI
  • Random Forest
  • Ridge Regression
  • SHAP

Infant Cry Recognition

  • Audio Analysis
  • Facial Analysis
  • Context-aware Prediction
  • Gradient Boosting
  • MediaPipe
  • XGBoost Fusion

Technology Stack

Frontend

  • React Native
  • Expo
  • TypeScript

Backend

  • FastAPI
  • Python

Database

  • Supabase

Machine Learning

  • TensorFlow
  • Scikit-learn
  • XGBoost
  • Random Forest
  • LSTM
  • SHAP

Cloud

  • Supabase

Repository Structure

TinySteps
│
├── backend/
├── frontend/
├── ml_models/
├── datasets/
├── assets/
│   ├── appICON.png
│   └── screenshots/
├── documentation/
└── README.md

Installation

Clone Repository

git clone https://github.com/SL-Predetor/infant-growth-monitoring-system.git

Backend

cd backend

pip install -r requirements.txt

uvicorn app.main:app --reload

Frontend

cd frontend

npm install

npx expo start

API Documentation

http://localhost:8000/docs

Research Team

Member Research Component
Yasindu Kularathne Multi-modal ASD Detection
Nethmi Sansala Premachandra AI Growth Monitoring
Dewmi Hennayake Maternal Recovery Assessment
Wikasith Dilshan Infant Cry & Discomfort Recognition

Supervisor

Ms. Gaya Thamali Dassanayake

Programme Coordinator
BSc (Hons) Data Science
Sri Lanka Institute of Information Technology


Project Resources

GitHub Repository

https://github.com/SL-Predetor/infant-growth-monitoring-system

Documentation

https://drive.google.com/drive/folders/122_Rl3Q7lMbTo35e0Tf5d9CL34YqQDOi


License

This project was developed for academic research purposes.


TinySteps
Empowering Caregivers Through Artificial Intelligence

About

TinySteps is an AI-powered mobile healthcare platform that leverages machine learning and explainable AI to support infant growth monitoring, developmental screening, maternal recovery assessment, and infant discomfort recognition. The platform delivers real-time, personalized insights to assist caregivers and healthcare professionals in making inf

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Used by

Contributors

Languages